A New Quality Measure for Image Segmentation Based on Combination of Information Redundancy and Variation of Information

This work presents a new combined measure for improving the quality of digital image segmentation. The measure has two components. The first is a measure of information redundancy, and the second is variation of information. Such a measure makes it possible to obtain an image partition that provides a compromise between the objectives of minimizing the number of selected informational important segments and minimizing the information dissimilarity between the original image and segmented image. A computational experiment conducted on a set of test images from the Berkeley University database confirmed the possibility of improving segmentation results when using a combined measure compared to the previously used measure of information redundancy.